TCBR Used In Experience Reuse Of Realty's Investment

The task of this dissertation is to study how to build a experience reusing system of realty's investment and exploitation using textual CBR technology and how to achieve the description of textual cases, the parsing of information entities, the retrieval and learning of cases in this system.CBR (Case-base reasoning) is one of AI reasoning methods and based on the cases. When a new problem is came up we abstract it into a new case and find a old similar case from the case base .So we can get some useful suggestions from the old case. If the old case is not fit to the new case, then we adapt the old one to fit the new condition. If this is successful the adapted case is stored into the case base and when the same condition comes up again we will get some useful information from this stored case. This is general CBR system's principle. It only does some work in the problem domains self-learning. But the experience of realty's investment and exploitation always has uncertain structure and its sources are many and varied. So we use textual CBR (a embranchment of CBR) to build case retrieval and learning module. Some new features come up and we will study, analyze and test them in this dissertation.In this dissertation, we first expatiated on the topic selection's background of the system and introduced the researching content and purport. From these description, we can see the using of TCBR and CRN technology in this system can help many investor as a intelligent assistant decision-maker and promote the study of the CBR and TCBR technology's applications. In succession, we recommend the conception and definition of TCBR, CRN and some key terms, and we expatiate on the building process of CRN and some correlative computing model and function. In the buildingprocess, we put forward a newly computing function of cases' correlative degree and make a particular analysis. Then we expatiate on the realization's arithmetic, structure and process of CRN in this system which include the parsing of information entities and the correlative degree's computing of cases.Case learning is the most important part of a CBR system. In this dissertation, we put forward a case learning method in TCBR and conceive its development. In case learning process, we come up with a feasible settle to solve the case redundant problem in this system.At the latter part of this dissertation, we discuss the advantage and limitation of this system including the comparison between existing reality assistant helper and our system, the comparison of performance and characteristic between traditional search based keyword and our system's CRN retrieval method based information entities. From those comparisons, we can know it is a feasible solution applying the TCBR and CRN technologies into this kind of assistant system.